WO1994005795A1 - Dna sequences encoding peach polygalacturonase - Google Patents

Abstract

The present invention relates to DNA sequences which encode the peach enzyme polygalacturonase. The present invention further relates to DNA constructs including these sequences and to recombinant plants of the family Rosaceae including these constructs. The DNA sequences are set out in Tables 1, 2 and 3.

Description

DNA SEQUENCES ENCODING PEACH POLYGALACTURONASE

The present invention relates to DNA sequences which encode the peach enzyme polygalacturonase. The present invention further relates to the use of these DNA

sequences in modulating polygalacturonase expression in peaches.

BACKGROUND OF THE INVENTION

Soft, juicy flesh is a feature of ripe peach fruit. In fruit of fresh eating varieties softening occurs in two stages. During early ripening, tissue firmness decreases slowly and progressively. Towards the end of ripening, loss of tissue firmness is rapid. This second stage of softening is called the "melting" stage. After "melting", fruit are very susceptible to physical injury and are stored for a few days only.

Fruit of peach varieties used for canning do not have a "melting" phase of softening. Ripe fruit remain relatively firm and maintain their shape throughout processing. These fruit have good keeping qualities but have not been generally favoured for fresh consumption because of their relatively dense flesh. Breeding projects that aim to produce a nonmelting peach desirable for fresh eating are currently underway (Sherman et al., 1990).

The "melting" flesh character is dominant and segregates in a Mendelian fashion (Bailey and French, 1949). It is tightly linked to the freestone character which is controlled by a single locus.

Fruit of "melting" varities show an increase in activity of endopolygalacturonase (EC 3.2.1.15, endoPG) during ripening (Pressey and Avants, 1978). EndoPG activity increases slightly during initial softening and markedly during the melting stage of softening in fruit of the freestone variety Flavorcrest (Orr and Brady, 1993). Significant endoPG activity has not been detected in "nonmelting" fruit (Pressey and Avants, 1978). Polygalacturonase is believed to contribute to fruit softening through its action on intercellular and cell wall pectins (Brady, 1987). In peaches, endoPG activity may be necessary for "melting" of fruit flesh to occur.

A number of polygalacturonase (PG) gene and cDNA sequences from plants have been described (DellaPenna et al., 1986, Gierson et al., 1986, Brown and Crouch, 1990, Lee et al., 1990, Niogret et al., 1991, Kutsumai et al., 1992, Atkinson and Gardner, 1993, Dopica et al., 1993, Lonsdale, 1993). Comparison at the amino acid level reveals a high level of conservation in certain regions of these sequences.

Expression of the peach gene (Lee et al., 1990) could not be demonstrated in fruit by methods that included the use of PCR (Lester, unpublished). Screening peach fruit cDNA libraries for ripening-related genes and for genes cross-hybridising with tomato endoPG cDNA failed to isolate a peach endoPG cDNA (Callahan et al., in press).

The present inventors have isolated and sequenced 3 cDNA sequences and a genomic DNA sequence which encode peach polygalacturonase enzymes. One of these cDNA sequences, PRF5 , encodes an endopolygalacturonase enzyme. The elucidation of these sequences enables the manipulation of these polygalacturonase enzymes in peach plants.

Accordingly, in a first aspect the present

invention consists in a DNA molecule which encodes a peach polygalacturonase enzyme, the DNA molecule having a sequence substantially as shown in Table 1, Table 2, Table 3 from residue 689 to residue 1526 or Table 3.

In a preferred embodiment of this aspect of the present invention the DNA molecule encodes a peach endopolygalacturonase enzyme and has a sequence

substantially as shown in Table 3. Observed homologies in the polygalacturonase sequences isolated from other plants suggest that the sequences isolated from peach will be similar to the polygalacturonase sequence of other members of the Rosaceae family. Accordingly, the sequences described herein will have use in other members of the Rosaceae family.

The sequences of the present invention can be used to manipulate plants of the Rosaceae family to either prevent expression of polygalacturonase enzymes present in the plant or, alternatively, to enhance expression of polygalacturonase enzymes in the plant.

In a second aspect the present invention consists in a DNA construct comprising at least 100 base pairs of the DNA molecule of the first aspect of the present invention, and a transcriptional initiation region functional in peaches or other members of the Rosaceae family, the at least 100 base pairs of the DNA molecule being joined 3' to the 3' terminus of the

transcriptional initiation region functional in peach or other members of the Rosaceae family.

In a third aspect the present invention consists in a DNA construct comprising the DNA molecule of the first aspect of the present invention, being joined 5' to the 3' terminus of a transcriptional initiation region functional in peach or other members of the Rosaceae family.

In a fourth aspect the present invention consists in a genetically engineered plant of the Rosaceae family, the plant containing the DNA construct of the second or third aspect of the present invention.

In addition to the use of antisense genes, the antisense genes may include sequence(s) coding for catalytic regions of riboendonucleases. In preferred forms of the invention the member of the Rosaceae family is prunus and most

preferably,peach.

These sequences also enable more efficient peach breeding programs to be conducted. Breeding programs are presently in progress in an effort to develop peach strains which possess the desirable characteristic of both "melting" and "non-melting" peach strains.

Progeny from breeding crosses can be assessed at an early stage for the presence of DNA which hybridises with the DNA molecules of the present invention. A hybridisation with these DNA sequences, particularly PRF5, is indicative that the progeny will produce

"melting" fruit.

Analysis of restriction fragment length

polymorphisms (RFLP) of the gene detected by PRF5 and the correlation of the RFLP's with phenotype will also provide a rapid means of determining whether a

particular plant will produce "melting" or "non-melting" fruit.

Accordingly, in a fifth aspect, the present

invention consists in a method of assessing whether a peach plant will produce "melting" fruit comprising determining if the plant includes a DNA sequence

corresponding to the sequence shown in Table 3.

As will be recognized by persons skilled in the art, the sequences of the present invention can be used to produce polygalacturonase enzyme using recombinant DNA technology.

Accordingly, in a sixth aspect, the present

invention consists in recombinant polygalacturonase enzyme, the polygalacturonase enzyme being coded for by the DNA molecule of the first aspect of the present invention.

In a seventh aspect, the present invention consists in a method of producing recombinant polygalacturonase enzyme, the method comprising transforming a cell with the DNA molecule of the first aspect of the present invention, culturing the cell under conditions which allow expression of the DNA molecule and recovering the polygalacturonase enzyme from the culture medium.

The cultured cell may be a bacterial cell, yeast, or other eukaryotic cell.

In order that the nature of the present invention may be more clearly understood, preferred forms thereof will now be described with reference to the following examples and accompanying drawings in which:- Fig. 1 Shows the restriction map of the 13.2 kb peach

DNA insert in the lambda PPG2 clone. Included in the figure is the location of the 3.5 kb fragment isolated in clone pPPG. The structure and orientation of the gene sequence identified within the fragments by sequencing, and the strategy used for sequence determination, are indicated. Restriction fragments subcloned for the purposes of sequencing are indicated with letters. Fragments A,C,F,H and J were subcloned using the restriction enzyme Rsa 1, sites of which are too numerous to indicate on the diagram. Other fragments are identified by the restriction enzymes used to generate them, for example KS17, is a Kpn/Sac fragment. Arrows without letters indicate sequence determined by priming with a synthetic oligonucleotide.

Double stranded sequencing of the original and subcloned fragments was carried out in both directions (indicated on the diagram) by oligopriming and chain termination (Sanger, Nicklen and Coulson, 1977). Synthetic oligonucleotides were used as primers to extend sequence determined by universal and reverse primers on the subcloned fragments. Restriction sites indicated are, Eco RI (E), Sac 1 (S), Hind 111 (H), Kpn 1 (K) and Xba 1(X). Xba mapping of the 5' end of the 13.2 kb fragment is

incomplete.

Fig. 2 Is a comparison of the exon/intron structures of the peach and tomato genes.

Fig. 3 Is a comparison of the N-terminal amino acid

sequence of the endopolygalacturonase enzyme from peach fruit, aligned with sequences from comparable regions of the mature tomato enzyme, and deduced sequences from the isolated peach gene and Oenothera cDNA.

MATERIALS AND METHODS

Plant Material

For isolation and characterization of the gene sequence peaches (Prunus persica L. Batsch) were grown at Arcadia, New South Wales and harvested at commercial maturity (firmness about 6kg measured with an 8 mm probe on an Effe-gi penetrometer). The fruit were covered in polyethylene film and allowed to ripen at 21°C. When soft (ca. 0.5 kg firmness) the fruit were peeled and quartered, and the mesocarp tissue was frozen in liquid nitrogen prior to storage at -80°C.

For the isolation of DNA, vegetative buds were taken when the emerging leaves were about 10 mm long and the tissue processed immediately.

Five peach varieties listed below were collected: Maravilla - semi freestone - leaves collected in mid July Flavorcrest - semi freestone - leaves collect in mid Sept.

Coronet - freestone - leaves collected in mid Sept.

Summerset - freestone - leaves collected in mid Sept Fragar - clingstone - leaves collected in mid Sept.

For isolation and characterization of the cDNA sequences fruit of the "melting" flesh cultivars

Flavorcrest and Fragar and the "nonmelting" cultivar Carolyn were used in these experiments. Flavorcrest fruit were collected from early ripening to commercial harvest stage from Arcadia, New South Wales (NSW).

Fragar fruit at commercial harvest stage were collected from Bathurst, NSW. Ripening Carolyn fruit were obtained from Stanhope, Queensland. Flesh firmness of fruit was measured by removing a small disc of skin from each side of the fruit and recording the force in kilograms

required to insert an Effegi penetrometer with an 1mm probe. For storage, the skin and seed of fruit were removed and the mesocarp was frozen in liquid nitrogen and placed at -80°C.

Measuring ethylene evolution of fruit

Fruit were collected at commercial harvest stage (approximately 6kg firmness), weighed and placed in sealed 800 mL jars ventilated with humidified air at flow rates of 1-1.2 L/h at 20°C. Ethylene concentrations were measured in duplicate lmL samples of air taken from outlet tubes of ventilated jars. Ethylene in the air was measured after separation on a Loenco model 15 CFX gas chromatograph fitted with a 1.52m × 0.32cm column of 60 to 80 mesh alumina and flame ionisation detection.

Ethylene production of Flavorcrest fruit was monitored daily for twelve days after harvest and three fruit with typical ethylene readings were sacrificed every two days for RNA analysis. Fragar and Carolyn fruit were

monitored until ethylene production peaked.

Enzyme extraction, isolation and assay

To assay endopolygalacturonase, the frozen fruit tissue was thawed in an excess of 1% (w/v) NaHSO₃, homogenized and centrifuged (13,000g, 20min). The deposit was washed with 1% (w/v) NaHSO₃ and then

extracted with 20 mol m^-3 solium acetate, 200 mol m^{- 3} NaCl, 10 mol m^-3 dithiothreitol pH 4.5 for 3 h at 2°C. The extract was recovered by centrifugation (20,000g, 20 min), concentrated by ultrafiltration (membrane retaining 30,000 M_r) and assayed by measuring the rate of change in specific viscosity (1/h^-1 g^{- 1}) during incubation at 40°C in 2% (w/v) polygalacturonic acid (Sigma) pH 4.3.

For purification of the enzyme, 1 kg of frozen tissue (var Coronet) was thawed in 50 mol m^-3 sodium lactate, 1000 mol m^-3 NaCl, 40 mol m^-3 b-mercaptoethanol pH 4.5 (1 dm³) and then homogenized. The homogenate was centrifuged (20,000g, 40 min) and the supernatant was reduced in volume by ultrafiltration (Amicon CH2

concentrator, S1Y30 cartridge), diluted 10 fold with

50 mol m^-3 sodium lactate pH 4.5, and again concentrated by ultrafiltration to 300 cm³. The concentrated extract was centrifuged (35,000 g, 40 min) and the supernatant pumped through a S-sephacryl FF (Pharmacia) column

(25 × 250 mm) equilibrated with 50 mol m^-3 sodium

lactate, 50 mol m^-3 NaCl pH 4.5. The column was washed with 2 column volumes of equilibration buffer and eluted with a linear gradient to 50 mol m^-3 solium lactate, 1 mol dm^-3 NaCl pH 4.5. Fractions (10 cm³) were assayed by incubating subsamples in 20 mol m^-3 sodium citrate, 0.2% polygalacturonic acid (Sigma) pH 4.5, at 40°C for 4 hr and measuring the increment in reducing potential by the method of Gross (1982). Fractions with significant activity were concentrated by ultracentrifugation and fractionated by FPLC using a Superose 12 (Pharmacia) gel permeation column in 20 mol m^-3 2-(N-morpholino)-ethano sulphonic acid (MES), 50 mol m^-3 NaCl pH 6.0. Active fractions were detected by assay and applied to a Mono S (HR 5/5) column equilibrated with the MES, NaCl pH 6.0 buffer. The Mono S column was developed with a linear gradient to 700 mol m^-3 NaCl in 20 mol m^-3 MES pH 6.0. Purification of peach endopolygalacturonase from fruit and sequence analysis

Pericarp tissue (800g) from ripe peach fruit,

Flavorcrest variety, was powdered in liquid nitrogen and then stirred into 100 mol m^-3 lactate, 200 mol m^-3 NaCl, 1 mol m^-3 dithiothreitol (DTT), 1% polyclar, pH4. After stirring for an hour at 2°C, the extraction was

centrifuged at 10,000 g for 1 hr and the supernatant filtered through two layers of miracloth. The filtrate was diluted 1:1 with 50 mol m^-3 lactate, 1 mol m^-3 DTT, pH4, and loaded onto a S-Sepharose FF (Pharmacia)

Cation-exchange column equilibrated with 50 mol^-3

lactate, 100 mol m^-3 NaCl, pH4. The column was washed with 50 mol m^-3 lactate, 50 mol m^-3 NaCl, 1 mol m^-3 DDT, pH4 and the protein was then eluted with a concave gradient to 1000 mol m^-3 NaCl in 50 mol m^-3 lactate, 1 mol m^-3 DTT, pH4. Peach endopolygalacturonase (PPG) eluted at approximately 400 mol m^-3 NaCl and active fractions were pooled and concentrated by ultrafiltration (Amicon Centricon 30) to 10 ml and loaded onto a

Conconavalin A - Sepharose 4B (Pharmacia) pectin affinity column equilibrated in 50 mol m^-3 lactate, 500 mol m^-3 NaCl, 1 mol m^-3 CaCl₂, 1 mol m^-3 MgCl₂, 1 mol m^-3 MnCl₂ pH4 (Buffer A). The column was washed with Buffer A containing 1 mol m^-3 DTT. PPG bound weakly to the column eluting with Buffer A. Active fractions were pooled and concentrated, diluted 1:20 with 25 mol m^-3 bis-Tris pH7.1, 1 mol m^-3 DTT (Buffer B) and reconcentrated to 2 ml. The sample was immediately loaded onto a Mono P (Pharmacia) chromatofocusing column equilibrated with

Buffer B. Protein was eluted by a linear pH gradient to pH5 with a 10% solution of Polybuffer 75 (Pharmacia). PPG eluted at pH 5.5 - 5.6. After elution, fractions were immediately titrated to pH4 with HCl and the active fractions were concentrated by ultrafiltration. On

SDS-PAGE (Chua and Bennoun, 1975), the Mono P- purified PPG ran as a single band at approximately M_r 45 . 000 .

After SDS-PAGE , PPG was transferred to PVDF membrane (Millipore) as described by Matsudaira (1987).

N-terminal protein sequence was determined on a Milligen 6600 Prosequencer (Millipore-Waters) using Sequelon-di-isothiocyante coupling reagent (Waters

Assc.).

Gel electrophoresis and immunoblotting

Tissue samples were ground in 2% (w/v) sodium dodecyl sulphate, 100 mol m^-3 β-mercaptoethanol, boiled for 3 min and clarified by centrifuging (20,000g, 10 min). Column fractions were precipitated with 10% (w/v) trichloracetic acid; the precipitate was boiled in 2% (w/v) SDS, 100 mol m^-3 β-mercaptoethanol and clarified by centrifugation. Proteins were fractionated by

SDS-polyacrylamide gel electrophoresis (Speirs & Brady, 1981) and blotted onto nitrocellulose (Kyhse-Andersen, 1984). The transferred polypeptides were reacted with immunoglobins specific for tomato fruit

endopolygalacturonase PG2A (Ali & Brady 1982) and the immunodetection was completed using peroxidase-coupled goat anti-rabbit immunoglobin (Towbin, Staehelin &

Gordon, 1979). Immunoglobins specific for the tomato PG2A enzyme were prepared using an affinity column with the purified protein coupled to Sepharose 4B (Pharmacia). DNA isolation

DNA was isolated from 10g batches of leaves

essentially by the method of Lichtenstein & Draper

(1985). The isolated DNA was purified by CsCl/ethidium bromide equilibrium centrifugation and the purified DNA stored at 4°C in TE buffer (10 mol m^-3 Tris/HCl pH 8, 1 mol m^-3 EDTA).

Restriction digestion

DNA samples were digested with restriction enzymes from Boehringer Manheim (Aust), according to the

manufacturers instructions.

Hybridization analysis of restriction fragments

Digested DNA was fractionated on 0.4% agarose gels and transferred to Zeta-Probe according to Zeta-Probe protocols (Bio-Rad Laboratories). Transferred DNA was hybridized with DNA probes oligolabelled with ³²P-dATP by the method of Feinberg & Vogelstein (1983).

Hybridizations were carried out in 4 × SSPE (SSPE is 1200 mol m^-3 NaCl, 40 mol m^-3 Na₂HPO₄, 4 mol m^-3 EDTA pH 7.7); 50% deionized formamide, 1% SDS, 1% skim milk solution (Diploma) and 10% dextran sulphate, essentially as described by Reed & Mann (1985). Hybridizations with homologous probes were for 20hr at 42°C and with

heterologous probes, 48hr at 37°C. After hybridization the filters were washed 2 × 10 min in 2 × SSC (SSC is 150 mol m^-3 NaCl, 15 mol m^-3 Na3 citrate), 0.5% SDS at room temperature followed by two 20 min washes in 0.1 × SSC, 0.05% SDS at 50°C. After blotting dry, filters were exposed to X-ray film (Fuji RX) at -80°C with two

intensifying screens for 1 to 5 days.

Cloning of peach genomic DNA

(i) Eco RI digested DNA

200mg of genomic DNA from Maravilla peach leaf was cut to completion with Eco RI restriction enzyme, the reaction stopped by heating at 70°C for 20 min and the DNA purified by partitioning against chlorofoπn:phenol (1:1) followed by chloroform at 4°C several times. The DNA was precipitated by the addition of an equal volume of 7.5 mol dm^-3 NH4 acetate and two volumes of ethanol and storage at -20°C overnight. The purified DNA was resuspended in 100 mol m^-3 Tris/HCl, 10 mol m^-3 MgCl₂, pH 8, and was fractionated on a 10 - 40% (v/v) glycerol gradient in the same buffer by centrifugation in a

Beckman SW41 rotor at 40,000 rpm for 15 hr, 15°C.

Fractions were collected from the base of the tube and aliquots were analysed by electrophoresis on a 0.4% agarose gel, Southern transfer (Southern, 1975) and hybridization, as described above, with ³²p-labelled tomato PG cDNA ( Sheehy et al, 1987). Fractions

containing DNA to which the cDNA hybridized were

identified and the DNA precipitated by the addition of three volumes of ethanol, storage overnight at -20°C and centrifugation at 18,000g for 40 min, 4°C. The pelleted DNA was resuspended in H₂O and aliquots were ligated into the plasmid vector pUC18 (Yanisch-Peron, Vieira and

Messing, 1985) using T4 ligase and overnight incubation at 4°C. The transformation of E.coli strain JM101 cells with ligated DNA fractions and the initial isolation of transformed cells containing peach DNA inserts, were carried out by standard procedures (cf Maniatis, Fritsch & Sambrook, 1982). Final screening of the transformed cells was undertaken by colony hybridization (Grunstein & Hogness, 1975) using ³²P-labelled tomato PG cDNA as the probe.

(ii) Partial Eco RI digested DNA.

200 mg total DNA from Maravilla peach leaf was digested for 2 hr at 37° with 0.03 units of Eco R1 enzyme per mg of DNA. The reaction was stopped, the DNA

cleaned, precipitated and resuspended as in (i) above, and was fractionated on a 10% - 40% glycerol gradient as above, but at 111,000 g for 16 hr, 15°. Fractions were collected from the base of the tube and aliquots analysed by agarose gel electrophoresis. Fractions containing DNA fragments with an approximate size of 10 - 15 kb, were further analysed by fractionation, transfer to

Zeta-Probed and hybridization with ³²P-labelled pPPG 1 insert (see text). The fraction containing DNA with homology to pPPG 1, was precipitated as before and was resuspended in a minimum volume of H2O. Aliquots were ligated into the bacteriophage vector EMBL4 (Frischauf et al, 1983), were packaged according to the protocol of

Scalenghe et al (1981) and were plated on E. coli strain TC 410 (Woodcock et al 1988). Plaque lifts were taken onto nitrocellulose membrane by standard procedures (Maniatis, Fritsch and Sambrook, 1982) and were screened by hybridization with ³²P-labelled pPPG 1 insert.

Plaques which hybridized with the probe were rescreened at low density, isolated and stored in dimethyl

sulphoxide (Maniatis, Fritsch and Sambrook, 1981) until required further.

Analysis of the peach 3.5 kb DNA fragment

To facilitate sequencing, the 3.5 kb fragment was digested with Rsa 1 restriction enzyme and was subcloned into the Sma 1 site of the plasmid pUC 18 (Yanisch-Peron, Viera and Messing, 1985). Double stranded sequencing in both directions was carried out on the original clone and on the Rsa 1 subclones using Sequenase (United States

Biochemical Corp.), universal and reverse M13 primers and oligonucleotide primers synthesised to specific regions of the clones. Overlapping sequence data (see Fig 5) was used to construct a composite map of the fragment.

Sequence analysis and comparisons were undertaken using computer programmes in the Cornell Package

(Fristenski, Lis and Wu, 1982).

Analysis of the lambda clone

A restriction map of lambda clone 2 was constructed by analysing single and double digest of the DNA using a number of restriction enzymes (see Fig. 1). Additional information was obtained by hybridization analysis of the fractionated, digested DNA samples with regions of the 3.5 kb insert of clone pPPG 1. Nucleotide sequence determination was carried out on the region of the cloned DNA fragment deemed to contain the peach

endopolyglacturonase gene. Double stranded sequencing of subcloned regions of the DNA was as described for the 3.5 kb fragment in the pPPG 1 clone (above).

RNA Preparation

RNA was extracted from fruit tissue by the following protocol which was based on that of Callahan et al.

(Callahan et al., 1989). Fresh frozen fruit mesocarp was powdered in a coffee grinder and 1 g was added to 20 mL of 100 mM Tris-HCl pH9, 100 mM NaCl, 1% SDS (w/v), 1% PVP-360 (w/v), 1% β-mercaptoethanol (v/v), 100 μg/mL Proteinase K (Boehringer Mannheim) and left to stand for five minutes at room temperature. Cellular debris was removed by centrifugation at 16,000g for 10 min. The solution was extracted with phenol equilibrated with 10 mM Tris HCl pH 7.5, then phenol:chloroform (1:1) and finally chloroform: isoamyl alcohol (24:1). The aqueous phase was placed on ice and 0.1 vol of 3 M NaAcetate pH 4.8, 0.01 vol of 10% SDS and 0.1 vol of 5 M NaCl were added. After incubation on ice for 2h, the precipitate was removed by centrifugation at 13,000g for 20 min. An equal volume of 6 m LiCl was added and the solution was incubated at 4°C overnight. The RNA was pelleted by centrifugation at 27,000g for 30 min and resuspended in 400 μL of water. The RNA was precipitated in 0.1 M NaCl with 2.5 vol of 95% ethanol and pelleted by

centrifugation. The RNA pellet was washed with 70% ethanol, dried at room temperature and pressure and resuspended in 100 μL of water. The concentrations of RNA solutions were calculated from absorption readings taken at 260 nm.

Total RNA was passed through a cellulose column to remove carbohydrates and then over oligo d(T) -cellulose for poly(A)⁺ + selection (Aviv and Leder, 1972).

Oligonucleotide sequence and design

The sequences of oligonucleotides (5'-3') used as PCR primers and the conserved PG amino acid sequences they matched were:

1. AAT/CACIGAT/CGGIA/GTICA NTDGIH

2. CCATGTCTTGATCCTAACTCC GVRIKTW

3. GGCGATGATTGCGTCTCTCTTG GDDCVSLG

Primers 1 and 3 are in the "sense" orientation.

Primer 2 is in the "antisense" orientation. The sequence of primer 4 was the same as the adaptor primer described in the protocol for "rapid amplification of cNDA ends"

(RACE, Frohman et al., 1988). Primer 1 included all the permutations of DNA sequence coding for the amino acid sequence given, with the nucleotide base of inosine used at positions that were totally degenerate. Primer 2 was synthesised as a sequencing primer for a peach PG genomic clone (Lee et al., 1990) and primer 3 was designed according to sequence of the PCR product of primers 1 and 2; these therefore do not show degeneracy.

First strand cDNA synthesis and PCR

First strand cDNA synthesis and PCR (Saiki et al., 1988) reactions are based on the RACE method (Frohman et al., 1988).

First strand cDNA was synthesised from 2 μg of poly (A) + RNA from Flavorcrest fruit at <0.5 kg firmness in 20 μL of IX PCR buffer with 1 mM dNTP's, 20 U rRNasin (Promega), and 4 U of AMV-RT (Promega) primed by an oligonucleotide with the sequence (5'-3')

GACTCGAGTCGACATCGA(T)₁₇.

PCR reactions were carried out in a volume of 50 μL with IX PCR buffer, 1.8 mM MgCl2, 0.4 mM dNTP's, 0.4 μM of primers, 10μL of the cDNA synthesis reaction and 1.25 U Taq polymerase (Perkin-Elmer-Cetus). Reactions were cycled 35 times at 94°C for 1 min, 55°C for 1 min and 72°C for 40 seconds.

PCR products were cloned into T-tailed (Marchuk et al., 1990) pBluescript KS vector (Stratagene) and

sequenced by the dideoxynucleotide method (Sanger at al., 1977) using Sequenase (United States Biochemical).

The sequence of a 255 bp fragment produced from PCR with primers 1 and 2 had similarity to previously

described PG sequences and was designated PRF1. PCR with primers 3 and 4 was designed to amplify the 3' cDNA end of PRF1, however this was unsuccessful and, instead, produced a distinct sequence, PRF3, which also showed PG sequence similarity. Lambda-ZAP cDNA library construction

A cDNA library was made from 5μg of poly(A) + RNA prepared from Flavorcreast soft (<0.5 kg) ripe fruit using a ZAP-cDNA synthesis kit (Stratagene) in the Lambda phage vector Uni-ZAP XR (Stratagene) according to methods described by the manufacturer. Phage was packaged in Gigapack Gold (Stratagene).

Library screening and analysis of positive clones

Duplicate plaque lifts of forty thousand primary recombinants on Biotrace NT membrane (Gelman) were screened with insert DNA prepared from PRF1 and PRF3.

Insert DNA was labelled to high activity with

[α-³²P]-dATP using a Bresatec random-labelling kit.

Hybridisation and washing of membranes was as recommended (Gelman) with radioactive probes at a concentration of 1 × 10⁶ cpm/mL. Positive plaques were purified to a single species and pBluescript KS phagemid DNA with insert was excised in the presence of helper phage, R408, and rescued as plasmid DNA according to Stratagene methods.

Positive cDNA clones with the largest inserts were sequenced at the 5' end. A longer version of the PRF3 cDNA, PRF5 comprising 1497 nucleotide base pairs and containing a complete open reading frame, was identified.

The complete sequence of both strands of PRF5 was obtained from nested deletions generated by the Erase-A-Base system (Promega).

Northern analysis

RNA gels were as described by Fourney (Fourney et al., 1988) with 10 μg of total RNA from each sample loaded. RNA was transferred to Zetaprobe membrane

(Biorad) by methods described by the manufacturer.

Filters were probed with [α-³²P]-ATP labelled insert DNA from PRF1 and PRF3. PRF3 was used as a probe, and not PRF5, because it was shorter and, therefore, less likely to hybridise nonspecifically. A cDNA with sequence related to ACC oxidase, also known as ethylene-forming enzyme, was isolated from a peach fruit cDNA library on the basis of cross

hybridisation with the tomato cDNA pT0M13 (Holdsworth et al., 1987). The coding sequence of this clone, PA01, was identical to that described by Callahan et al., 1992.

Insert DNA of PA01 was used as a probe in northern analysis.

Sequence comparison

Similarities between the derived amino acid

sequences of PRF1, PRF3 and previously described PG sequences were determined using the FASTA program

(Pearson and Lipman, 1988) and the Genbank database.

RFLP Analysis

For RFLP analysis, genomic DNA was prepared from the melting flesh "Springcrest" and the non-melting flesh "Carolyn" varieties, by the method of Thomas et al., (1993).

10μg of DNA was digested with 30 units of EcoRl at 37°C for 2 hours and electrophoresed through a 1% agarose gel in running buffer of 40mM Tris/HCl pH 7.8, ImM EDTA and 5mM sodium acetate. The DNA was then transferred to Zetaprobe membrane (BioRad) in 0.4m NaOH, for 4 hours. Filters were hybridised at 65°C with ³²p-labelled insert DNA of PRF3 and washed at high stringency using methods described for the membrane (BioRad). Washed filters were exposed to X-ray film with an intensifying screen, overnight at -80°C.

RESULTS

Enzyme activity

No activity was detected in assays of mature fruit at harvest. In ripe, soft fruit of cultivars Coronet and Flavorcrest (midseason, freestone cultivars) the activity was 8.5 ± 0.3 × 10^-5 (6) 1/ η min ^-1 g^-1 for the 1987/88 season and 8.3 + 0.4 × 10^-5 (4) 1/ η min^-1 g^-1 for the 1988/89 season. This activity is 10^-4 to 10^-5 that in ripe tomatoes (Brady et al 1982; Brady et al, 1983).

Fractionation of the enzyme

At pH 4.5 in the presence of 50 mol m^-3 NaCl, the endopolygalacturonase activity was retained by the cation exchange resin S-Sepharose FF, permitting the separation of the enzyme from the viscous polysaccharides in the crude extracts. When eluted from the cation exchange column, the concentrated extract was fractionated by gel permeation chromatography (FPLC, Sepharose 12). The activity eluted in a fraction of M_r 45,000 ± 10,000. At 20 mol cm^-3 MES pH 6.0, it was fractionated on a cation exchange column in a NaCl gradient. The activity was separated from residual exo-PG activity (Downs & Brady, 1990), and was on the leading edge of a major protein peak. It was confirmed as an endo-enzyme by its activity in viscosity assays and by its inhibition by calcium (Pressey & Avants, 1978). These preliminary attempts to fractionate the protein indicates that it is a relatively basic protein (binds strongly to Mono S at pH 6.0), of M_r in the vicinity of 45,000.

Immunoblotting

Extracts of peaches at harvest and when soft ripe, were fractionated by SDS-polyacrylamide gel

electrophoresis, blotted onto nitrocellulose and

challenged with antisera to tomato polygalacturonase PG2A. A strong cross reaction to a M_r 45,000 polypeptide was detected in soft ripe but not in mature, unripe fruit. The size of the reactive band corresponded with the size of the tomato PG 2A polypeptide and the reaction in ripe but not unripe fruit paralleled the distribution of endopolygalacturonase activity. When fractions from the Mono S column used to purify the enzyme were

separated by SDS-gel electrophoresis and immunoblotted, the strongest antigenic activity was in the fractions with most enzyme activity. It seemed possible that the tomato PG-specific immunoglobulins were detecting

homologous epitopes in the peach endopolygalacturonase protein.

The activity of the tomato enzyme is strongly inhibited in the presence of specific antiserum. Adding the tomato anti-PG serum to extracts containing the peach enzyme, significantly reduced their activity.

Non-specific serum did not inhibit the enzyme (Table 5).

The enzyme inhibition is further evidence that the tomato and peach enzymes have regions of sequence homologies.

Table 5 Inhibition of endopolygalacturonase activity by specific antiserum.

Treatment Activity % of control

Tomato enzyme, control 100

Tomato enzyme, non-specific serum 104

Tomato enzyme, specific serum 6

No enzyme, ± serum 0

Peach enzyme, control 100

Peach enzyme, non-specific serum 90

Peach enzyme, specific serum 48

Enzyme was incubated in buffer or 1 mg cm^-3 specific or non-specific serum for 1 hr at 40°C, before substrate was added. Incubation with substrate was for 15 min

(tomato) or 16 hr (peach). Figures are percentages of controls for tomato and peach enzyme respectively.

Treatments were in duplicate.

Endopolygalacturonase enzyme from peach fruit

Peach endopolygalacturonase was isolated from 800 g of fruit and N-terminal amino acid sequence determined as described in Materials and Methods. The sequence is indicated in Fig. 3.

The peach endopolygalacturonase gene

The immunological evidence of sequence homologies between the tomato and peach endopolygalacturonase proteins suggests that there is some degree of homology between the tomato and peach genes. To investigate this, peach genomic DNA was digested with Eco RI, Sac 1 and Hind 111 restriction enzymes, fractionated by agarose gel electrophoresis and probed by hybridization with

³²P-labelled cDNA to tomato endoPG. At low stringency, the tomato cDNA hybridized predominantly with peach DNA fragments of sizes 3.5 kb, 6.4 kb and 1.9 kb

respectively, indicating sequence homology.

To further investigate the extent of homology, peach DNA was digested with EcoR1 enzyme, size fractionated on a glycerol gradient and fractions containing DNA

fragments of approximately 3.5 kb were collected and purified. These were ligated into the plasmid vector pUC 18, and transformed E.coli cells were screened with

³²P-labelled cDNA to tomato endoPG. A plasmid pPPG 1 was isolated containing a 3.5 kb insert of peach DNA which hybridized at low stringency with the tomato endoPG cDNA.

To examine the possibility of isolating a larger fragment of genomic DNA with regions bounding the 3.5 kb fragment, peach DNA was partially digested with Eco RI, fractionated by agarose gel electrophoresis and

transfered to Zeta Probe. Hybridization with

3²P-labelled 3.5 kb fragment revealed a ladder of larger DNA fragments with homology to the probe, and presumably containing the 3.5 kb fragment and adjacent regions of DNA. A 200 mg preparation of DNA was partially digested with Eco RI under the same conditions and was

fractionated by glycerol gradient centrifugation. DNA with a size of approximately 13.5 kb, equivalent to one of the larger hybridizing bands, was isolated from the gradient and purified. The DNA fraction was ligated into lambda EMBL 4 arms (Frischoff et al, 1981), packaged and plated on E. coli. 20,000 plaques were screened by hybridization with ³²P-labelled 3.5 kb DNA fragment and 12 positive colonies were identified and isolated. One of the colonies, lambda PPG2, was further characterized by restriction analysis and sequencing.

Characterization and sequencing of the 3.5 kb fragment and the lambda PPG2 clone

2.9 kb of the 3.5 kb peach DNA fragment were

subcloned and sequenced as described in Materials and Methods and in Fig. 1. Comparison of the nucleotide sequence and its derived amino acid sequence with those of the published tomato endoPG gene (Bird et al, 1988), revealed various regions of homology between the

sequences. The homologous regions corresponded to exons 5 to 9 of the tomato gene indicating that the 3.5 kb peach fragment represents the 3' end of the peach endoPG gene but lacks the 5' end.

A restriction map of the 13.5 kb peach DNA fragment in the lambda PPG2 clone is shown in Fig. 1. The

location of the 3.5 kb pPPG1 fragment within the 13.5 kb fragment is indicated in the figure and was confirmed by hybridization of the 3.5 kb fragment to fractionated digests of the lambda PPG2 DNA.

Regions of the 13.5 kb DNA fragment were subcloned and sequenced as described in Materials and Methods.

Within the fragment is a putative gene sequence (Table 1) of ca 2.5 kb with 8 exons and 7 introns, and with well delineated exon/intron boundaries. There are two ATG codons in exon 1 but the more 5' of the two has been indicated as the initiation codon on the basis of its position upstream of the other and on the basis that there is an A at position -3 to the ATG in the nucleotide sequence, making this a stronger initiation codon than the ATG at +30 (Kozak, 1984; Joshi, 1987). Upstream of the ATG at position 1, are a transcription start site CTCATCC at position -8 to -14, which is very similar to the concensus initiation start site for plant sequences, CTCATCA, described by Joshi (1987), with a potential

'TATA' box, TATAATT, at position -21 to -27. At the 3' end of the gene, a perfect poly A addition concensus sequence AAATAA is located 67 bases downstream of the stop codon.

Comparison of the 8 exons of the peach gene with the 9 exons of the tomato endoPG gene indicates 69% homology at the nucleotide level and 54% homology at the level of deduced amino acid sequence, within the region defined as the processed peptide in the tomato (Sheehy et al, 1987), identifying the peach gene as encoding polygalacturonase. The difference in length of the two genes is due

primarily to differences in intron sizes and to the absence of one intron in the peach sequence (Fig. 2).

The coding regions of the peach gene have been deduced from exon/intron boundaries, open reading frames and by comparison with the sequence of the tomato

endopolygalacturonase enzyme (Sheehy et al, 1987; Bird et al, 1986). The gene encodes a peptide of 407 amino acids in length. In Table 4, the deduced sequence of the peach peptide is compared with the amino acid sequences of tomato (Bird et al, 1986) and Oenothera (Brown and

Crouch, 1990) endopolygalacturonases. The peach sequence is 54% conserved with respect to the mature tomato sequence and 33% conserved with respect to the same region of the Oenothera sequence. Of particular interest are two regions of high conservation in all three

sequences, boxes 1 and 2. In box 1, the conserved area flanks a central histidine residue, which may represent the active site of the enzyme. Located close to this putative active site, is box 2 which may be associated with substrate binding. Of the 14 cysteines in the tomato peptide, nine are conserved in both peach and Oenothera peptides and two of the remainder are

transposed by one or two amino acids, suggesting

conservation of secondary structure derived from

disulphide bridges. Isolation of partial PG-related cDNAs

The nucleotide and derived amino acid sequences of the 255 bp PCR product of primers 1 and 2, PRF1, are presented in Table 2. Comparison of PRF1 with Genbank sequences resulted in a highest match with Oenothera organensis polygalacturonase with 66% identity over 80 residues. The sequence of the 800bp PCR product, PRF3, from primers 3 and 4 was distinct from that of PRF1 but also showed PG sequence similarity.

Construction and screening of Flavorcrest Lambda ZAP cDNA library

A total of 120,000 primary recombinants were

obtained in the Lambda-ZAP cDNA library of which over ninety percent contained inserts. The screen with PRF1 produced no positive plaques. The screen with PRF3 produced approximately 400 positive plaques. One clone, PRF5, with an insert of 1526 bp, was selected for

sequence analysis.

Sequence analysis of PRF5

The nucleotide and derived amino acid sequences of

PRF5 are presented in Table 3. Sequence analysis of PRF5 revealed an open reading frame that codes for a

polypeptide of 393 amino acids with a predicted molecular mass of 41,500 D. Highly conserved regions of PG

sequence were present in the sequence. Comparison of

PRF5 with GenBank sequences resulted in a highest match with kiwifruit polygalacturonase showing 41% overall identity.

The N-terminal sequence of the mature endoPG protein isolated from ripe peach fruit was determined to be

TPVTYNVASLGAKADGKTDST7AFLS. A corresponding sequence was found in the amino acid sequence of the predicted protein encoded by PRF5, twenty three amino acids from the putative translation commencement point. (see Table 3 and Fig.3. Analysis of RNA related to PRF1, PRF3 and PAO1

The advance of softening was one parameter against which RNAs in peach fruit related to PRF1, PRF3 and PAOl were studied. Northern analysis detected no RNA hybridising with PRF1 in Flavorcrest, Fragar and

Carolyn fruit at any stage of softening. PCR using primers 1 and 2 amplified a product with the same size as PRF1 in cDNA prepared from Flavorcrest fruit of 4 and <0.5 kg firmness but not in fruit of >12 or 6 kg firmness. The possibility that PRF1 represented contaminating peach genomic DNA sequence was ruled out because PCR with the same primers on a template of genomic DNA gave a product of a different size (500 bp). Southern analysis confirmed that PRF1 represents sequence in the peach genome (data not shown).

In unripe Flavorcrest fruit of firmness >12 kg no RNA hybridising with PRF3 was detected by northern analysis. A PRF3-related sequence of 1.7 kb was

detected in fruit of 6 and 4 kg firmness when filters were exposed for 3 days. Results from exposure of filters for 24 h showed that a massive increase in PRF3-related RNA occurred as fruit lost tissue firmness from 4 kg to <0.5kg.

PAOl-related RNA was present as a 1.6 kb

transcript at low levels in fruit of 12 kg firmness and at higher levels in fruit of 6 and 4 kg firmness. It was abundant in ripe fruit of less than 0.5 kg

firmness.

Levels of RNA related to PAO1 and PRF3 during ripening were studied according to the parameter of ethylene evolution by Flavorcrest fruit during a postharvest period. PAOl-related RNA was present at low levels when ethylene evolution from fruit was

<10 μL kg^-1 hr^-1. In these fruit, PRF3 detected RNA only after three days exposure of filters. for both

PRF3 and PAO1, a sudden increase in related RNA levels occurred in the one or two days before the climacteric peak associated with the final stages of ripening and remained high for four days afterwards. "Melting" of fruit flesh occurred in the days when these increases were observed.

In the Fragar and Carolyn fruit, levels of RNA that hybridised with PRF3 were found to peak around the climacteric (data not shown) . Fragar fruit underwent "melting", but, when ripe, were much firmer than

Flavorcrest fruit. Limited sensitivity of the

penetrometer meant that this difference was not

quantitated, with both varieties of ripe fruit

registering <0.5 kg. Ripe fruit of the nonmelting variety Carolyn were firm relative to both Fragar and Flavorcrest, with a penetrometer reading of 6 kg.

Comparison of climacteric fruit of the different varieties showed much lower levels of 1.7 kb RNA detected by PRF3 in Fragar than those in Flavorcrest. In Carolyn fruit PRF3 detected levels of mRNA below those found in Flavorcrest and Fragar. There was an obvious difference in size of the hybridising

transcript in Carolyn compared to that in Flavorcrest and Fragar. It was estimated to be 250 bp shorter. RFLP Analysis

In Springcreast DNA, PRF3 (PRF5) detected 3 hybridising fragments of approximate sizes 3.6, 1.3 and 0.6 kg. In Carolyn DNA the probe detected only one fragment of 3.6 kb. Thus it appears that PRF3 (5) may be useful for distinguishing the "melting flesh" and "non-melting flesh" varieties by RFLP analysis.

PAO1 RNA was present as a 1.6 kb transcript in climacteric fruit of all three varieties. Carolyn had the highest levels of PAO1 message and Fragar had levels slightly lower than Flavorcrest. There was no relationship between relative levels of ACC oxidaserelated RNA in the three varieties and their relative ethylene readings. Ethylene readings for the

Flavorcrest, Fragar and Carolyn fruit were 40, 55 and 65 μL kg^.1hr^.1., respectively.

DISCUSSION

In freestone peaches, there is a particularly rapid loss of firmness and increase in "water soluble pectin", that is related to an increase in both exo- and

endo-polygalacturonase activities (Pressey, Hinton and Avants, 1971; Pressey and Avants, 1978). Because a similar increase in exopolygalacturonase activity occurs in clingstone peaches which do not soften rapidly or show a significant increase in soluble pectin (Pressey and Avants, 1978) there is reason to suggest that

endopolygalacturonase plays a decisive role in the softening of peach fruit. Softening may also be

associated with de-esterification of the pectins (Reeve, 1959), although it is not clear whether the decrease in methyl esterification that occurs with ripening is a prelude to endopolygalacturonase activity or a

consequence of wall disruption. If endopolygalacturonase is the key catalyst of softening in peaches, it is remarkably efficient, for the softening is achieved with from 0.01 to 1.0 per cent of the activity that is found in ripe tomato fruit.

Sufficient peach enzyme was purified to allow the determination of the N-terminal amino acid sequence. The sequence is shown in Fig. 3 in comparison with sequences from the same regions from other PG molecules. There is a conserved region NVxSxGAKGDGKT with respect to the N-terminal of the mature tomato enzyme which confirms the identity of the peach peptide.

Despite the differences in the apparent efficiencies with which the peach and tomato enzymes contribute to fruit softening, there is sufficient homology in the two systems for the peach protein to be recognized by

polyclonal antibodies to the tomato enzyme, and for the gene to be selected by hybridization to a portion of the tomato gene. The specificity of the immunological cross-reactivity was established by the lack of reaction with proteins from mature, non-ripening peach fruit, by the coincidence of antigen and enzyme activity when the enzyme was fractionated for purification, and by the inhibition of enzyme activity by specific, but not by non-specific, sera.

Comparison of the putative peach

endopolygalacturonase gene with endopolygalacturonase genes from tomato and Oenothera shows significant

homology in the coding regions and some interesting structural similarities and differences. The unprocessed peptide encoded by the peach gene is 407 amino acids in length compared to 456 amino acids encoded by the tomato gene. Major differences are at the amino terminus which appears to be 35 amino acids shorter than the tomato peptide, and the carboxyl terminus which is 14 amino acids shorter than the tomato peptide. Interestingly, 13 amino acids are removed from the carboxyl terminal of the tomato peptide during post-translational processing

(Sheehy et al, 1987). The absence of these 13 amino acids in the peach peptide suggests that it may not be subjected to C terminal processing.

Charged amino acids in the processed "mature" tomato peptide, give the peptide slightly basic properties.

Amino acid differences between the same regions in the peach and tomato peptides result in the peach peptide being considerably more basic than the tomato peptide, by seven +ve charges. Similarly, the mature tomato peptide has four potential N-linkage, glycosylation sites, three of which are believed to be glycosylated in vivo (Sheehy et al, 1987) while the same region of the peach peptide has only two potential linkage sites, with neither conserved in relation to the tomato peptide. It

therefore appears that, despite a significant degree of conservation at both gene and peptide level, the peach peptide differs in N- and C-termini, net charge and

N-glycosylation.

Comparison of the N-terminal amino acid sequence of the purified peach enzyme with the equivalent region deduced from the isolated peach gene (Fig. 3) shows limited homology between the sequences indicating a relationship but clear evidence that the enzyme was not encoded by the characterized gene. Indeed, the amino acid sequence of the enzyme more closely resembles the sequences of the tomato and Oenothera peptide N-termini (Fig. 3).

The gene sequence was obtained from the peach variety Maravilla while the sequenced enzyme was from another semi-freestone variety, Flavorcrest. It is unlikely that the noted differences in sequences result from varietal differences, and more likely that the results imply the presence in peach of more than one gene for endopolygalacturonase. This has been demonstrated to be the situation in Oenothera (Brown and Crouch, 1990) but has not been defined in tomato.

A cDNA, PRF5, isolated from ripe Flavorcrest peach fruit was shown to encode the endoPG protein found in Flavorcrest fruit during ripening. This conclusion was based primarily on the finding of a match between part of the deduced amino acid sequence of PRF5 and N-terminal sequence of the purified protein (Figure 3).

In tomato, the PG cDNA, pTOM6 , was positively identified as encoding the endoPG protein found in fruit during ripening (Grierson et al., 1986). Direct

correlation of other PG sequences with protein products has not been demonstrated. The conclusion of Niogret et al. (Niogret et al., 1991) that the maize cDNA sequences code for an exo-form of the enzyme (EC 3.2.1.67) is not substantiated by experimental evidence. The unprocessed peach protein predicted from PRF5 nucleotide sequence was 393 amino acids long and had a molecular mass of 41,500 D. It contained one potential N-glycosylation site at a similar position to one found in tomato endoPG.

The putative leader sequence of peach endoPG was 23 amino acids long which was much shorter than the 70 amino acid sequence of tomato endoPG. Characteristic features of a leader sequence, including hydrophobic and hydrophilic regions were present in these 23 amino acids (von Heijne, 1983). From PRF5 sequence, the predicted molecular mass of endoPG after processing is 39,000 D, which compares with a M_r of 44,000 reported for the native protein.

The accumulation of RNA that hybridised to a partial clone of PRF5, PRF3, in relation to softening of Flavorcrest fruit occurred in a similar pattern to increases in endoPG activity (Orr and Brady, 1993). Low levels of both were associated with the gradual softening of early ripening. A marked increase in both PRF3-related RNA and endoPG activity was associated with the "melting" stage of softening. There was a relationship between levels of RNA that hybridised to PRF3 and the degree of softness of ripe fruit between cultivars in Flavorcrest, Fragar and Carolyn.

The smaller RNA transcript detected by PRF3 in fruit of the "nonmelting" variety Carolyn is of particular interest. Significant endoPG activity was not detected in "nonmelting" fruit, which led to the theory that absence of the enzyme accounted for the "nonmelting" phenotype (Pressey and Avants, 1978).

From northern analysis of Carolyn fruit, it appears that transcription of the endoPG gene may still occur in "nonmelting" fruit. The smaller size of the RNA transcript in Carolyn fruit may reflect a sequence aberration that affects translation and production of active enzyme.

Polygalacturonase genes have been described from tomato (Bird et al., 1988), maize (Allen and Lonsdale, 1992), kiwifruit (Atkinson and Gardner, 1993) and peach (Lee et al., 1990) and cDNA sequences form tomato fruit (Dellapenna et al., 1986, Grierson et al., 1986), maize pollen (Niogret et al., 1991, Rogers et al., 1991), Oenothera organenεis pollen (Brown and Crouch, 1990), avocado fruit (Kutsumai et al., 1992, Dopico et al., 1993) and tobacco pollen (Lonsdale, 1993).

Pollen-specific polygalacturonase clones from maize and Oenothera organensis form small multigene families with relatively high conservation among coding sequences within the species of >99% and 89%,

respectively, (Brown and Crouch, 1990, Allen and

Lonsdale, 1992. Comparison of overlapping regions of the deduced amino acid sequences of PRF1. PRF3 and the genomic PG clone PPPGl reveal that there is

considerable divergence between PG sequences of peach. This divergence may reflect differences in primary role of the polygalacturonases produced by the three genes. Genes represented by PRF1 and PRF5 are both expressed in fruit but the gene encoding PRF5 appears to have a more significant role. PRF5 RNA is far more abundant than PRF1 and appears to code for the prevalent endoPG enzyme in fruit. PRF1 RNA is associated with ripening and present at very low levels. The role of the gene represented by PRF1 is not clear. Expression of the peach PG gene represented by the genomic clone PPPG1 could not be demonstrated in fruit, using sensitive methods. Expression of this gene may take place in other tissues, such as pollen or abscission zones. REFERENCES

Ali, Z. M. & Brady, C. J. (1982) Purification and characterization of the polygalacturonases of tomato fruit; Australian Journal of Plant Physiology, 9, 155-169.

Allen RL, Lonsdale DM (1992) Sequence analysis of three members of the maize polygalacturonase gene family expressed during pollen development. Plant Mol Biol

20: 343-345.

Atkinson RG, Gardner RC (1993) A polygalacturonase gene from kiwifruit. Genbank Accession number L12019.

Aviv H, Leder P (1972) Purification of biologically active globin messenger RNA by chromotography on oligothymidylic acid-cellulose. Proc Natl Acad Sci USA 69: 1408-1412.

Bailey JS, French HP (1949) The inheritance of certain fruit and foliage characters in peach. Mass Agr Expt

Sta Bui 452.

Bird CR, Smith CJS, Ray JA, Moureau P, Bevan MJ, Birds AS, Hughs S, Morris PC, Grierson D, Schuch W (1988) The tomato polygalacturonase gene and ripening specific expression in transgenic plants. Plant Mol Biol 11:

651-662.

Brady, C. J., MacAlpine, G., McGlasson, W. B. & Ueda, Y. (1982) Polygalacturonase in tomato fruits and the induction of ripening. Australian Journal of Plant

Physiology, 9, 171-178.

Brady, C.J., Meldrum, S.K. & McGlasson, W.B. (1983).

Differential accumulation of the molecular forms of polygalacturonases in tomato mutants. J. Food

Biochem., 7, 7-14.

Brady CJ (1987) Fruit ripening. Ann Rev Plant Physiol

38: 155-178.

Brown SM, Crouch ML (1990) Characterization of a gene family abundantly expressed in Oenthera organensis pollen that shows sequence similarity to

polygalacturonase. Plant Cell 2: 263-274.

Callahan A, Morgens P, Walton E (1989) Isolation and in vitro translation of RNAs from developing peach fruit. Hort Sci 24: 356-358.

Callahan AM, Morgens PH, Wright P, Nichols KE (1992) Comparison of Pch313 (pTOM13 homolog) RNA accumulation during fruit softening and wounding of two

phenotypically different peach cultivars., Plant

Physiol 100: 482-488.

Callahan A.Morgens PH, Cohen R (in press) Isolation and initial characterization of cDNAs for mRNAs regulated during peach fruit development. J Amer Soc Hort Sci. Dellapenna D, Alexander DC, Bennett AB (1986) Molecular cloning of tomato fruit polygalacturonase. Analysis of polygalacturonase mRNA levels during ripening. Proc Natl Acad Sci USA 83: 6420-6424.

Dopioco B, Lowe AL, Wilson ID, Merodio C, Grierson D (1993) Cloning and characterization of avocado fruit mRNAs and their expression during ripening and low-temperature storage. Plant Mol Biol 21: 437-449.

Downs, C.G. & Brady, C.J. (1990) Two forms of

endopolygalacturonase increase as peach fruit ripen. Plant, Cell and Environment 13, 523-530.

Feinberg, A.P. & Vogelstein, B. (1983). A technique for radiolabeling DNA restriction endonuclease

fragments to high specific activity. Anal. Biochem., 132, 6-13.

Fourney RM, Miyakoshi J, Day RS, Paterson MC (1988) Northern blotting: efficient RNA staining and

transfer. Focus 10: 5-7.

Frischauf, A-M. , Lehrach, H., Poustka, A. and

Murray, N. (1983). Lambda replacement vectors carrying polylinkler sequences. J. Mol. Biol. 170, 827-342 Fristensky, B., Lis, J. & Wu, R. (1982) Portable microcomputer software for nucleotide sequence

analysis. Nucleic Acid Research, 10, 6451-6463.

Frohman MA, Dush MK, Martin GR (1988) Rapid

amplification of full-length cDNA ends from rare transcripts: Amplification using a single gene-specific oligonucleotide primer. Proc Natl Acad Sci USA 85:8998-9002.

Grierson D, Tucker GA, Keen J, Ray J, Bird CR, Schuch W (1986 Sequencing and identification of a cDNA clone for tomato polygalacturonase Nuci Acids Res 14: 8595-8603. Gross, K. (1982) A rapid and sensitive method for assaying polygalacturonase using 2-cyanoacetamide.

Hortscience, 17, 933-934.

Grunstein, H. and Hogness, D.S. (1975) Colony

hybridization, a method for the isolation of cloned DNAs that contain a specific gene. Proceeding National Academy Science U.S.A., 72, 3961-3965.

Holdsworth MJ, Bird CR, Ray J, Schuch W, Grierson D (1987) Structure and expression of an ethylene-related mRNA from tomato. Nucl; Acids Res 15: 731-739.

Joshi, C.P. (1987). An inspection of the domain between putative TATA box and translation start site in 79 plant genes. Nucleic Acids Research, L5, No 16, 6643-6653.

Kozak, M. (1984). Compilation and analysis of

sequences upstream of the translation start site in eukaryotic mRNAs. Nucleic Acids Research, 12, 857-872 Krieger BL, Doleman MS (1988) Use of an alternative reaction vessel and support media for protein sequence analysis. J Protein Chem 7:255.

Kutsumai SY, Lin A-C, Percival FW, Laties GG,

Christoffersen RE (1992) Avocado polygalacturonase mRNA. Genbank Accession number L06094.

Kyhse-Andersen, J. (1984) Electroblotting of multiple gels: a simple apparatus without buffer tank for rapid transfer of proteins from acrylamide to nitrocellulose. Journal of Biochemical and Biophysical Methods, 10, 203-209.

Lonsdale DM (1993) Tobacco pollen-specific

polygalacturonase mRNA Genbank Accession number X71019. Lee E, Speirs J, Gray J, Brady CJ (1990) Homologies to the tomato endopolygalacturonase in the peach genome. Plant, Cell Environ 13: 513-521.

Lichtenstein, C.P. & Draper J. (1985) Genetic

engineering of plants. In DNA Cloning, vol.2, a practical approach Ed. D.M. Glover, IRL Press, Oxford and Washington, P.P. 67-119. Maniatis, T., Fritsch, E.F. & Sambrook, J. (1982) Molecular Cloning: A

Laboratory Manual. Cold Spring Harbor Laboratory, Cold Spring Harbor, NY.

Marchuk D, Drumm M, Saulino A, Collins FS (1991)

Construction of T-vectors, a rapid and general system for direct cloning of unmodified PCR products. Nucl Acids Res 19:1154.

Maniatis, T., Fritsch, E.F. & Sambrook, J. (1982).

Molecular Cloning: A Laboratory Manual. Cold Spring Harbor Laboratory, Cold Spring Harbor, NY.

Matsudaira P (1987) Sequence from picomole quantities of proteins electroblotted onto

polyvinyldienedifluoride membrane. J Biol Chem 262: 1035-1038.

Niogret M, Dubald M, Mandaron P, Mache R (1991)

Characterization of pollen polygalacturonase encoded by several cDNA clones in maize. Plant Mol Biol 17: 1155-1164.

Orr G, Brady C (in press) How endopolygalacturonase activ relates to fruit softening in a dessert peach. Postharv Biol Techn.

Pearson WR, Lipman DJ (1988) Improved tools for

biological sequence comparison Proc Natl Acad Sci USA 85: 2444-2448. Pressey, R., Hinton, D.M. & Avants, KJ.K. (1971).

Development of polygalacturonase activity and

solubilization of pectin in peaches during ripening. J. Food Sci., 36, 1070-1073.

Pressey R, Avants JK (1978) Differences in

polygalacturonase composition of clingstone and

freestone peaches. J Food Sci 43: 1415-1423.

Reed, K.C. & Mann, D.A. (1985) Rapid transfer of DNA from agarose gel, Nucleic Acid Research, 13, 7207-7221. Reeve, R.M. (1959). Histological and histochemical changes in developing and ripening peaches. II. The cell walls and pectins. Amer. J. Botany, 46, 241-248. Rogers HJ, Allen RL, Hamilton WDO, Lonsdale DM (1991) Pollen specific cDNA clones from Zea mays. Biochim Biophys Acta 1089: 411-413.

Saiki RK, Scharf S, Faloona F, Mullis KB, Horn GT, Erlich H, Arnheim N (1985) Enzymatic amplification of β-globin genomic sequences and restriction site

analysis for diagnosis of sickle cell anaemia. Science 230: 1350-1354.

Sanger F, Niklen S, Coulson AR (1977) DNA sequencing with chain terminating inhibitors. Proc Natl Acad Sci USA 74 5463-5467.

Sheehy, R.E., Pearson, J., Brady, CJ. & Hiatt, W.R. (1987) Molecular characterization of tomato fruit polygalacturonase. Molecular and General Genetics, 208, 30-36.

Sherman WB, Topp BL, Lyrene PM (1990) Non-melting flesh for fresh market peaches. Proc Fla State Hort Soc 103: 293-294.

Southern, E.M. (1975). Detection of specific sequences among DNA fragments separated by gel electrophoresis. Journal Molecular Biology, 98, 503-517.

Speirs, J. & Brady CJ. (1981) A co-ordinated decline in the synthesis of sub-units of ribulose bisphosphate carboxylase in aging wheat leaves. II abundance of messenger RNA. Australian Journal of Plant Physiology, 8, 603-618.

Thomas, M.R., Matsumoto, S., Cain, P. and Scott, N.S. (1993). Repetitive DNA of grapevine: classes present and sequences suitable for cultivar identification.

Theor.Appl. Genet. 86, 173-180.

Towbin, H., Staehelin, T. & Gordon, J. (1979)

Electrophoresis transfer of proteins from

polyacrylamide gels to nitrocellulose sheets: procedure and some applications. Proceedings of the National Academy of Sciences, U.S.A., 76, 4350-4354.

Von Heijne G (1983) Pattern of amino acids near signal-sequence cleavage sites. Eur J Biochem 133: 17-21

Woodcock, D.M., Crowther, P.J., Diver, W.P.,

Graham, M., Bateman, C, Baker, D.V. and Smith, S.S.

(1988). RglB facilitated cloning of highly methylated eukaryotic DNA: The human L1 transposon, plant DNA, and DNA methylated in vitro with human DNA methylase. Nucleic Acids Research 16., No 10, 4465-4482.

Yanisch-Peron, C, Vieira, J. & Messing, J. (1985) Improved MIS phage cloning vectors and host strain, nucleotide sequences of the M13 mpl3 and pUC19 vectors. Gene, 33, 103-119.

Claims

CLAIMS : -

1. A DNA molecule which encodes a peach

polygalacturonase enzyme, the DNA molecule having a sequence substantially as shown in Table 1, Table 2, Table 3, from residue 689 to residue 1526 or Table 3.

2. A DNA molecule as claimed in claim 1 in which the DNA molecule encodes a peach endopolygalacturonase enzyme and has a sequence substantially as shown in Table 3.

3. A DNA construct comprising at least 100 base pairs of the DNA molecule as claimed in claim 1 or claim 2 and a transcriptional initiation region functional in peach or other members of the Rosaceae family, the at least 100 base pairs of the DNA molecule being joined 3' to the 3' terminus of the transcriptional initiation region functional in peach or other members of the Rosaceae family.

4. A DNA construct comprising the DNA molecule of the first aspect of the present invention, being joined 5' to the 3' terminus of a transcriptional initiation region functional in peach or other members of the Rosaceae family.

5. A genetically engineered peach plant, the peach plant containing the DNA construct as claimed in claim 3 or claim 4.

6. A genetically engineered plant as claimed in claim 5 in which the plant is prunus.

7. A genetically engineered plant as claimed in claim 6 in which the plant is peach.

8. A method of assessing whether a peach plant will produce "melting" fruit comprising determining if the plant includes a DNA sequence corresponding to the sequence shown in Table 3.

9. A recombinant polygalacturonase enzyme, the polygalacturonase enzyme being coded for by the DNA molecule as claimed in claim 1.

10. A method of producing recombinant

polygalacturonase enzyme, the method comprising transforming a cell with the DNA molecule of claim 1, culturing the cell under conditions which allow expression of the DNA molecule and recovering the polygalacturonase enzyme from the culture medium.

AMENDED CLAIMS

[received by the International Bureau on 7 January 1994 (07.01.94); original claim 1 amended; other claims unchanged (1 page)]

1. A DNA molecule which encodes a peach

polygalacturonase enzyme, the DNA molecule having a sequence substantially as shown in Table 1, Table 2, Table 3 from residue 689 to residue 1496 or Table 3.

2. A DNA molecule as claimed in claim 1 in which the DNA molecule encodes a peach endopolygalacturonase enzyme and has a sequence substantially as shown in Table 3.

3. A DNA construct comprising at least 100 base pairs of the DNA molecule as claimed in claim 1 or claim 2 and a transcriptional initiation region functional in peach or other members of the Rosaceae family, the at least 100 base pairs of the DNA molecule being joined 3' to the 3' terminus of the transcriptional initiation region functional in peach or other members of the Rosaceae family.

4. A DNA construct comprising the DNA molecule of the first aspect of the present invention, being joined 5' to the 3' terminus of a transcriptional initiation region functional in peach or other members of the Rosaceae family.

5. A genetically engineered peach plant, the peach plant containing the DNA construct as claimed in claim 3 or claim 4.

6. A genetically engineered plant as claimed in claim 5 in which the plant is prunus.

7. A genetically engineered plant as claimed in claim 6 in which the plant is peach.

8. A method of assessing whether a peach plant will produce "melting" fruit comprising determining if the plant includes a DNA sequence corresponding to the sequence shown in Table 3.

9. A recombinant polygalacturonase enzyme, the polygalacturonase enzyme being coded for by the DNA molecule as claimed in claim 1. STATEMENT UNDER ARTICLE 19

The amendment removes an error contained in the definition of the sequence fragment given at line 4 of claim 1 as filed. That is claim 1 as filed provides that the sequence fragment is from "residue 689 to residue 1526" of Table 3. Clearly, Table 3 does not include 1526 residues. Instead, Table 3 indicates that the sequence comprises 1497 residue. This number is also in error as only 1496 residues are depicted.